banking market liberalization and bank performance: the ... · ownership, as denoted by different...
TRANSCRIPT
THE WILLIAM DAVIDSON INSTITUTE AT THE UNIVERSITY OF MICHIGAN
Banking Market Liberalization and Bank Performance:
the Role of Entry Modes
By: Ngoc-Anh Vo Thi
William Davidson Institute Working Paper Number 948 January 2009
-1-
Banking Market Liberalization and Bank Performance: the Role of Entry Modes
Ngoc-Anh Vo Thi
University of Paris X at Nanterre
200, Avenue de la République
92001 Nanterre Cedex, France
This version: August 2008. Comments welcome.
Abstract
This paper analyzes the evolution in bank performance following the removal of legal
restrictions on the entry of foreign banks in three transition economies: the Czech Republic,
Hungary, and Poland. Two modes of foreign bank entry are considered: entry by Greenfield
investments, and by foreign mergers and acquisitions of domestic banks. For this purpose, we
construct a panel data of banks from the three countries over the period 1994-2004. We
determine the dates on which liberalization occurred in each country. Bank performance is
reflected by accounting measures of profitability, net interest margin, and operating costs.
The results show a very limited effect of the entry of Greenfield banks on domestic banking
market in the early transition period. In contrast, the foreign entry by mergers and
acquisitions of domestic banks exerts significant impacts on bank performance. Indeed, we
observe significant declines in banks' profits and net interest margins, and a significant
increase in operating costs. Our results have important policy implications for those emerging
and transition economies still hesitant to liberalize their banking markets.
JEL Classification: G21, F36, C01.
Keywords: Banking, Transition Economies, Foreign Bank Entry, Greenfield Investment,
Mergers and Acquisitions, Bank Performance.
-2-
1. Introduction Over the past two decades, the world of banking has experienced an unprecedented trend
towards internationalization. One aspect of this increasing financial integration is the
dramatic change in foreign participation in emerging markets' banking systems, especially
since the second half of the 1990s. Central and Eastern Europe is a region that has seen a
particularly stunning increase in international banking presence. Indeed, foreign presence
here rose from around 10% at the end of 1994 to around 50% of total banking system assets
at the end of 1999 (Mathieson & Schinasi, 2000), and are now major players in these
markets.
There were two waves of foreign bank entry into these economies – first, by setting
up de novo investments in the host countries, and second, by acquiring (generally sizable
state-owned) domestic banking institutions. These entries have been made possible by
gradual relaxation of restrictions on both modes of entry by foreign banks. Given one of the
main objectives of these countries in liberalizing their banking systems is to promote banking
efficiency, it is useful to assess whether this policy has been effective.
So far, the most commonly adopted approach to assess the effect of foreign bank
entry on bank performance is to conduct a comparative analysis of bank performance
between foreign and domestic banks. Findings for emerging and transition economies
generally show that foreign banks fare significantly better than their domestic counterparts
even when different types of performance measures are taken into account. These findings
are taken to represent positive effects of foreign bank entry on bank performance. In our
opinion, this type of evidence is, to a certain degree, questionable, because the better
performance of foreign banks may raise the average performance of the whole banking
system, but this does not mean that foreign bank entry has helped improving the performance
of domestic banks. Yet, the objectives of governments in liberalizing foreign entry are not
only to admit good performers to their markets, but also to provide domestic banks with a
means to improve their own performance.
A more direct test on the effects of foreign bank entry on bank performance has been
to investigate the relationship between the change in the degree of foreign bank presence and
host markets’ bank performance. This relationship has been found to be positive and
significant, which has been interpreted as another evidence on the benefits that foreign banks
bring to host countries’ banking markets.
-3-
In this paper, we propose another direct approach for testing the effects of foreign
bank entry on bank performance by using a variance of the event study methodology. Indeed,
we investigate whether bank performance changes following the removal of legal restrictions
separately on each mode of entry of foreign banks, i.e. greenfield investments versus foreign
mergers and acquisitions of domestic banks1. To the best of our knowledge, this represents
the first attempt to do so in the context of transition economies. Moreover and most
importantly, we are the firsts to focus on the differential impact of the two modes of foreign
entry on host countries in an empirical setting.
The results show a very limited effect of the entry of Greenfield banks on domestic
banking market in the early transition period. In contrast, the foreign entry by mergers and
acquisitions of domestic banks exerts significant impacts on bank performance. Indeed, we
observe significant declines in banks' profits and net interest margins, and a significant
increase in operating costs.
Our paper is related to two strands of literature: the theoretical literature on foreign
bank entry, and the empirical literature on the effects of foreign ownership on bank
performance.
Theoretical analysis on foreign bank entry has been scarce. (Buch C. M., 2003) sets
up a theoretical model on foreign bank entry and demonstrates empirically that large
information barriers discourage the entry of foreign banks. (Hauswald & Marquez, 2003)
shows that information spillovers from incumbent banks to potential entrants lead to a
decrease in interest rates and profitability of banks. (Kaas, 2004) constructs a model of spatial
loan competition and concludes that foreign bank entry is generally too low in comparison
with the socially optimum level. (Sengupta, 2007) considers the consequences of asymmetric
information on foreign bank entry, and demonstrates that both better information ex-ante and
better creditor protection ex-post facilitates the entry of outside competitors with a cost
advantage.
Of particular interest to our paper are the three studies that have taken into account the
impact of the entry modes of foreign banks on host countries’ banking markets. Indeed,
(Claeys & Hainz, 2006) investigates the impact of entry modes on local credit markets, in
particular on domestic banks’ interest rates. It has been shown that de novo entry intensifies
1 We define foreign banks as those in which foreign owners (Companies + Individuals) hold at least 50% of total share capital. A greenfield investment involves the establishment of an institution from scratch, whereas a merger and acquisition implies the purchase of a firm's (here a bank's) shares or other form of capital. In this paper, we are exclusively interested in a control acquisition, i.e. the purchase of equal to or more than 50% of a bank's capital.
-4-
competition, and consequently domestic banks’ interest rates decrease. (Van Tassel &
Vishwasrao, 2007) studies entry mode choices by foreign banks, and their impact on host
countries. The results show that foreign banks prefer de novo entry to acquisition entry in
order to capture target banks’ informational advantages. On the other hand, it is most likely
that credit allocations will improve when foreign banks enter by acquiring target banks with a
relatively high proportion of good clients. With regard to de novo entry, this intensifies
competition, and consequently leads to a decrease in market interest rates. (Lehner &
Schnitzer, 2008) finds that de novo entry intensifies competition, and consequently leads to a
decrease of banks’ interest rates and market shares. This lessens the incentives of banks to
investigate in screening technology. Thus, the authors conclude that domestic banks’
incentives to invest in screening technology are stronger with acquisition entry than with de
novo entry. Our paper contributes to this literature by providing empirical evidence on the
differential impact of entry modes by foreign banks on bank performance.
Our paper is directly related to the empirical literature addressing the effect of foreign
ownership on bank performance. There are two main approaches to the topic. The first one is
to directly consider the impact of a change in regulatory conditions or a change in the level of
foreign bank participation on bank performance. General results indicate that foreign bank
entry is associated with an increase of competition in domestic banking markets, which leads
to an improvement in bank performance.2 The United States, Latin America, and Asia are
geographical areas that this line of literature has focused on3. No study has been carried out
on transition economies using this approach.
The second approach is to conduct a comparative analysis of bank performance
between foreign and domestic banks. The methodology used within this approach is on the
whole similar. Most of them first calculate the indicators of bank performance based on data
on financial statements of individual banks. Typically, three sets of indicators are used to
measure bank performance. First are standard measures of financial performance, namely
Return on Assets (ROA) and Return on Equity (ROE). Second are accounting measures of
operating performance and third are efficiency and productivity measures constructed
through frontier estimation techniques or data envelopment analysis. Then, these studies
regress these indicators on a set of explanatory variables, especially ownership variables, in
order to assess the importance of various determinants of bank performance. A limited 2 See (Barajas et al., 2000) (Claessens et al., 2001) (Denizer, 2000) (Unite & Sullivan, 2003) amongst others. 3 See (Chong, 1991) (Schranz, 1993) (DeYoung et al., 1998) (Jayaratne & Strahan, 1998) for US, (Ghosh et al., 2008) (Denizer, 2000) (Isik & Hassan, 2003) (Unite & Sullivan, 2003) for Asia, (Barajas et al., 2000) (Clarke et al., 1999) for Latin America, and (Claessens et al., 2001) (Bayraktar & Wang, 2004) for cross-country studies.
-5-
number of studies have taken into account the impact of the modes of entry of foreign banks
on bank performance.
Most geographical areas have been taken into account: United States, Western
Europe, Latin America, Eastern Europe in transition, and Asia including China. The results
are contradictory for developed and developing economies. Indeed, in developed economies,
foreign banks are found to be less efficient than their domestic peers. However, some banking
organizations, particularly from the United States, are found to consistently operate at or
above the efficiency levels of domestic banks (Berger, DeYoung, Genay, & Udell, 2000). By
contrast, empirical evidence on developing economies points out that foreign banks are more
efficient than domestic banks. In particular, in line with the broader case, studies focusing on
transition economies also find a positive relationship between foreign involvement, foreign
ownership, as denoted by different levels of participation of foreign banks in domestic banks,
and bank performance (Bonin, Hasan, & Wachtel, 2005b). Moreover, with respect to the role
played by the modes of entry of foreign banks, while Greenfield banks are found to
outperform both M&As and domestic banks, the comparative performance between M&As
and domestic banks remains unclear (Havrylchyk, 2006) (Vo Thi & Vencappa, 2008).
Thus, we contribute to this literature by directly considering the impact of foreign
bank entry on bank performance using an event study methodology. To the best of our
knowledge, this represents the first attempt to do so in the context of transition economies.
Moreover and most importantly, we are the firsts to focus on the differential impact of the
two modes of foreign entry on host countries in an empirical setting. Indeed, theoretical
literature on foreign direct investment is mostly concerned by the determinants of enterprises'
choice of entry modes. Little is said on their implications for host countries. This paper
therefore represents a step towards analyzing the welfare effects of different modes of FDI
into banking.
Our paper is organized as follows. Section 2 gives a brief account of banking sector
reforms and the associated entry of foreign banks into the Czech Republic, Hungary, and
Poland. Section 3 describes the data and the variables used in the empirical analysis. Section
4 discusses about the results. Conclusion follows.
-6-
2. Banking Sector Reforms and the Entry of Foreign Banks The aim of this section is to give the reader an account of banking sector reforms and the
associated liberalization to foreign bank entry in the three countries that serve as field studies
for this paper: the Czech Republic, Hungary, and Poland. We will draw on major themes and
discuss common and divergent features for these countries.
2.1. Banking at the beginning of the transition, and first wave of foreign bank entry by Greenfield investments
As is the case for other former communist economies, in the Czech Republic, Hungary, and
Poland, the transition towards a market-based banking system starts with the break-up of the
mono-bank system into a two-tier banking system with the former mono-bank assuming
central bank functions, and its separated entities assuming commercial bank functions. The
number and types of units into which the commercial operations of the mono-banks were
divided vary across countries.
Czech Republic In the Czech Republic, the break up led to the creation of one bank, Komercní Banka, which
inherited a loan portfolio consisting of almost all of the Czech Republic’s commercial loans
as well as the Czech component of the mono-bank’s payments system. Moreover, there are
four specialized banks: Cˇeska´ Sporˇitelna (CS), the savings bank; Investicd nıB Banka (IB)
specialized in development and investments involving foreign companies; Cˇeskoslovenska´
ObchodnıB Banka (CˇSOB), the foreign trade bank; and Zˇ ivnostenska´ Banka (Zˇ B)
specialized in foreign currency deposits.
Hungary In Hungary, three nation-wide banks were created (the Hungarian Credit Bank - MHB, the
Commercial and Credit Bank - K&H, and Budapest Bank - BB) and specialized by sectors,
i.e., industry, agriculture, and small business and utilities. In addition, there are four
specialized banks: the National Savings Bank (OTP), the Hungarian Foreign Trade Bank
(MKB), the General Banking & Trust (ÁÉB), and Postabank which was established by
government initiative in 1988 (Majnoni, Shankar & Várhegyi, 2003).
-7-
Poland In Poland, the new nine commercial banks are specialized by region, and effectively have
control of the region’s financial services in the short run. In addition, there are five
specialized banks: Bank Handlowy w Warszawie S.A. (BH), the foreign trade merchant
bank; Bank Polska Kasa Opieki S.A. (Pekao), a savings bank specialized in collecting retail
deposits in foreign currency and handling foreign currency transactions involving private
persons; Powszechna Kasa Oszcze¸dnos´ci- Bank Pan´stwowy (PKO BP), another savings
bank specialized in zloty household deposit-taking and in financing housing construction;
Bank Gospodarki Z° ywnos´ciowej S.A. (BGZ), a national umbrella bank for local
cooperative banks involved in the financing of agriculture and food processing; and Bank
Rozwoju Eksportu S.A. (BRE) specialized in export financing (Abarbanell & Bonin, 1997).
Each of the above approaches to carving out commercial banks from the former mono-bank
represents a trade-off between short-run franchise value derived from monopoly power and
long-run franchise value derived from efficiency in the banks. In fact, the minimal
restructuring in the Czech Republic and Hungary endowed the commercial banks with strong
market positions, thus enhancing their financial stability in the early reform period. However,
the absence of real competition does not provide these banks with the incentives to increase
their efficiency in the longer run term. The more extensive restructuring in Poland generated
banks with less market power and with portfolios more vulnerable to the economic conditions
in their regions. While some of these banks became insolvent or need government
intervention, others developed stronger portfolios, contributing to faster improvement in the
provision of banking services (Meyendorff & Snyder, 1997).
In parallel with the break-up of the mono-bank, banking acts were adopted allowing
the entry of new banks. Together with private domestic banks, foreign banks have taken this
opportunity to enter these countries. At this stage, their entries often take the form of
Greenfield investments.
Czech Republic Indeed, in the Czech Republic, the Act on Banks was passed in February 1992 which allowed
the licensing of foreign banks. Between 1989 and 1993, several foreign-owned banks were
set up, all specialized in investment banking and services to companies and high-revenue
households. They were created as either subsidiaries or branches of foreign banks (Weill,
2003).
-8-
Hungary In Hungary, before the start of the transition in 1987, foreign banks were already present in
the forms of offshore banks or joint-ventures with domestic banks. In the new era, liberal
bank licensing policies encouraged new bank entry in the market. Most of the de novo entries
of foreign banks are via wholly-owned subsidiaries. Moreover, foreign owners have bought
out the Hungarian partners in the early joint ventures. The Banking Act of December 1991
allowed foreign banks to have more than 10% of equity share in domestic banks.
Poland In Poland, the Act on Banking voted in January 1989 effectively opened up the market to
foreign investors. During the first two years of transition (1990−1991), four foreign banking
institutions were established of which three under their brand names, namely Raiffeisen-
Centrobank (established together by Raiffeisen Zentralbank Osterreich AG and Centro
Internationale Handelsbank AG), Creditanstalt and Citibank. ING Bank N.V. and Societe
Generale established branches in Warsaw. Seven other foreign banks were established in the
years 1990−1993 by a number of other foreign banks, investment funds, foreign companies
and, in some cases, with a small participation of Polish state-owned banks or enterprises and
state agencies (Balcerowicz & Bratkowski, 2001).
2.2. Banking crises and restructuring
Like in other transition countries, most of the state-owned commercial banks (SOCBs) in the
Czech Republic, Hungary, and Poland started with extensive non-performing loans. A larger
part of these troubled loans were inherited from the past communism. The other part was
created during the transition process due to the legacy of non-market-based allocations of
credits. By 1992, bad loans were estimated to vary from 2.4-19% to 50-66% of total loans in
the Czech Republic, to approximate 20.7% in Hungary, and 30% in Poland4 (Tang, Zoli, &
Klytchnikova, 2000), thus constituting one of the most critical obstacles to the operation of
the banking systems in the three countries.
In order to deal with these solvency crises, a number of actions have been taken to
clean up banks’ balance sheets, whose nature and frequency vary.
Czech Republic In the Czech Republic, in 1991, a consolidation program was initiated to clean up state-
owned banks’ balance sheets and to strengthen their capital. The program included the
4 These numbers should be taken as indicative.
-9-
following operations: (i) transfer of assets to Konsolidační Banka, a state institution aimed at
acquiring, administering, and amortizing the four state-owned banks’ communist-era bad
loans; (ii) write-off of loss loans from the National Property Fund’s resources; (iii) increases
in the capital of the banks split off from the State Bank of Czechoslovakia and of
Ceskoslovenska Obchodni Banka (CSOB); and (iv) transfer of credits and guarantees from
ČSOB to Česká Inkasní (ČI), a state institution created in 1993 to take over debt obligations
of foreign trade companies to facilitate privatization of CSOB. The net cost of this program
has been estimated at about CZK 100 billion (Tuma, 2002). The results were not satisfactory
as non-performing loans continued to be consistently high throughout the 1990s due to
persistent unsound lending practices and because of external events such as the May 1997
Krone crisis, the floods and investments in Russian derivatives (Bonin & Wachtel, 1999).
There are two more consolidation and stabilization programs in 1996-1997 to address
liquidity crises in small private banks. Moreover, a so-called London approach to loan
recovery was also adopted, consisting of a set of rules for reaching out-of-court settlements
between the banks and debtor enterprises.
Hungary In Hungary, in 1991, the Hungarian government formally guaranteed about half of the stock
of recognized non-performing loans in the three major state-owned commercial banks. Then,
in 1992, with a significant growth in new bad loans, a loan consolidation program was
initiated, aiming at transferring substantial portions of the commercial banks’ non-performing
loans to the Hungarian Development and Investment Company in exchange for government
bonds. Next, in the following year, this program was expanded both to encourage enterprise
restructuring and to promote the sale by banks of their bad debts to companies that
specialized in collections and workouts. As part of the program, the banks received nearly Ft
80 billion in state funds for recapitalization of the banks (Meyendorff & Snyder, 1997).
Moreover, in the case of MHB, the “good bank/ bad bank” approach was used as its bad
assets were transferred to a subsidiary, Risk Kft in july 1995, which was to be wounded up in
three years while the good bank was successfully privatized. The second recapitalization
ultimately proved to be successful because soon afterwards Hungary adopted a policy of
privatizing state banks by selling controlling shares to strategic foreign investors.
Nevertheless, there are two more recapitalization rounds to address solvency problems in the
state-owned agricultural bank (Agrobank) in 1995, and in the second largest retail bank
(Postabank) in 1997.
-10-
Poland In Poland, from 1989 to 1992, a significant part of the original bad loans were swept by
Poland’s hyper-inflation. After plans for restructuring enterprises and privatizing banks were
adopted, an Enterprise and Bank Restructuring Program was launched with an aim to
recapitalize and resolve the problem of nonperforming loans prior to the privatization of
state-owned banks. Using the Polish Bank Privatization Fund supported by the World Bank,
the Polish government decided to proceed with a one-time recapitalization of the banks,
based on the value of the portfolio of bad debts at the end of 1991. However, as a condition
for participating in the program, banks were obliged to create work-out units and actively
pursue collection through several instruments, and were required to resolve all their non-
performing loans at the end of 1994. Nine state-owned banks were affected by this reform.
The outcome of the restructuring program is generally considered as positive: the share of
non-performing loans in loan portfolios of the eleven major banks was reduced from 30% in
1992 to 9 percent in 1996 (Meyendorff & Snyder, 1997) (Weill, 2003).
Thus, the measures adopted to resolve bad loans problem in the Czech Republic and
Hungary represent a combination of a centralized approach (i.e. transferring of bad debts to a
“loan hospital”) and a decentralized one (i.e. non-performing loans remaining on the bank's
books, and work-out units being created within the banks to pursue the recovery of these non-
performing loans). In contrast, Poland adopted solely a decentralized approach. Moreover, in
Hungary and the Czech Republic, banking crises resolution was characterized by repeated
rounds of recapitalization, which calls into question the government’s commitment not to bail
out failed banks, and setting up moral hazard problems. In contrast, Poland has pursued a
more credible no-bailout policy with its one-time recapitalization program. The end results
were more successful for Hungary and Poland as the shares of troubled loans declined
subsequently to restructuring programs. By contrast, in the Czech Republic, these were
consistently high throughout the 1990s as noted above.
2.3. The Need for Privatization: Second Wave of Foreign Bank Entry by Acquiring Former State-Owned Banks
However, removing troubled loans from banks and recapitalizing them were far from enough
to improve bank efficiency and prevent recurrence of other banking crises, so far as banks
continue to be dependent on the government and its directed lending policies, and to
perpetuate links with poorly performing state-owned enterprises. The Czech experience has
shown that early recapitalizations were of little benefits because soft lending practices
-11-
continued and future bailouts became necessary. By contrast, the repeated recapitalizations in
Hungary were ultimately successful because privatization to an independent, usually foreign,
owner followed rapidly. The point here is that following recapitalization and restructuring,
the corporate governance problem in banks was to be addressed so that an independent
banking sector was developed, freed from non market-based practices. The key to a corporate
governance reform is to transfer bank ownership to private hands. Effectively, this was these
privatization processes that drew a second wave of foreign bank entries through mergers and
acquisitions of domestic banks.
Czech Republic Indeed, the Czech Republic was involved in two waves of privatisation of state-owned banks.
The large Czech banks were transformed into joint-stock companies in 1992 and three out of
the largest four state-owned banks were partially privatized within the first wave of voucher
privatization. CSOB was excluded because of its unique involvement in foreign financial
markets5. The state, nevertheless, kept controlling stakes in these banks, in line with the state
applied principles adopted in 1991 according to which the state would retain control of at
least 40-50% of the basic capital, foreign participation would be held to a 25% maximum,
and no single foreign investor would be permitted more than a 10% stake (Simonson, 2001).
The exception was Zivnostenka Banka, which was sold to foreign investors in 1992 (Weill,
2003) (Pruteanu-Podpiera, Weill, & Schobert, 2008).
Inadequate governance and lack of effective corporate restructuring led to a fragile
banking system with relatively low foreign participation. This meant that the first wave of
privatization proved to be a failure in resolving structural problems of banks. By 1998,
Czech government stakes were 65.7% in CSOB, 45% in CS, 36.3% in IPB, and 48.7% in KB
(Simonson, 2001, p. 201). The weaknesses of the state-owned banks with large holdings of
non-performing loans, and the goal of accession to the European Union has pushed the newly
elected government to adopt a new privatization program in April 1997 with the aim of
accelerating state-controlled banks’ privatization. In 1998, the government sells its stakes in
IPB to Normura (Bekaert & Harvey, 2004). The privatization process gathered speed in the
second half of 1999 with the sale of CSOB to Belgium’s second largest. In February 2000,
Austria’s second-largest banks agreed to buy a 52% stake in CS, bringing foreign
participation to about 54% (Mathieson & Schinasi, 2000). Also in 2000, Komercni Banka
was successfully privatized to Societe Generale of France. 5 The government feared that voucher privatization might alter CSOB’s viability and have negative effects on monetary stability (Simonson, 2001, p. 201).
-12-
Hungary In Hungary, the privatization program was launched in 1994 following two recapitalization
schemes. The first significant privatization took place with the partial sale of Hungarian
Foreign Trade Bank (MKB) in the summer of 1994 to EBRD (financial investor with a share
of 16, 68%) and Bayerische Landesbank (strategic investor with a 25,01% share) whose stake
increased two years later (in January 1996) when the state sold its 25 percent share. The same
scheme characterized the sale of Budapest Bank in 1995 to GE Capital (27.5% of shares) and
the European Bank for Reconstruction and Development (EBRD) (32.5%). During 1996-97
MHB, K&H and other commercial banks were sold to strategic foreign investors, according
to the basic scheme of the Hungarian bank privatization that privileged the pursuance of a
strategic foreign presence. A second scheme was followed by the Hungarian authorities with
the partial sale in the summer of 1995 of the largest retail bank, National Savings and
Commercial Bank (OTP). The transaction involved a private placement of Global Depositary
Receipts representing approximately 20% ownership in OTP in international capital market
and a domestic placement of approximately 15%. The government retained approximately
50% ownership stake and asserts control through the Board (Kormendi & Schnatterly, 1996).
By forbidding the presence of strategic investors and by promoting a prevalent domestic
ownership, the Government wished to create a diversified proprietary structure, dominated by
institutional investors. Overall, the privatization of the Hungarian banking system was
practically completed by the end of 1997. By that time state ownership had dropped to 21 per
cent of bank capital while the foreign stake had increased to over 60%. At the end of 2000
state ownership had dropped to 19% while the foreign stake had increased to over 66%
(Majnoni, Shankar, & Varhegyi, 2003).
Poland In Poland, the whole privatization process can be divided into two phases. The first phase,
from 1993 to 1997, is the one of delayed privatization, with foreign investors being allowed
to participate, but entitled only to minority shares. The second phase, from 1997 to 2000, can
be characterized as fast privatization, with all state-owned banks being privatized, most of
them to foreign strategic investors.
On March 16, 1993, the first step towards privatizing Wielkopolski Bank Kredytowy
(WBK) was carried out with an announcement on an issue of new shares to obtain a capital
injection of 28.5% of the augmented share base. EBRD purchased the entire new issue due to
absence of interest from a foreign strategic investor. The State Treasury retained a vast share
in equity (44.3%). In March 1995, Allied Irish Bank (AIB) participated in a second new share
-13-
issue by WBK and acquired a 16.26% stake. The new issue reduced EBRD’s stake to 24%
and AIB signed an agreement with the EBRD to purchase its stake at a future date. Later AIB
increased its stake to 36% by purchasing an additional 20% of the shares from the Treasury’s
holding (Abarbanell & Bonin, 1997). In 1996, AIB increased its stake to 60.2% in WBK by
exercising its option to buy the EBRD’s stake (Balcerowicz & Bratkowski, 2001).
In June 1993, a second commercial bank, Bank Slaski (BSK) was privatized. After
some difficulties, a strategic foreign investor, ING, has been attracted to the bank with a
25.92% stake. The State Treasury retained 33.16% in equity. In July 1996, ING purchased
additional shares from the state to bring its total shareholdings to 54.1%, the government
retained only a 5% stake.
In January 1995, the third commercial bank, Bank Przemyslowo-Handlowy (BPH)
was put to sale exclusively in a public offer. Due to a limited demand, the EBRD as the
major underwriter of the IPO was left with a 15.6% ownership stake. ING and BSK each took
5% of the shares, a stake that the group increased subsequently to 12%. The state retained a
47% stake in BPH after privatization while the remainder of the shares are widely held
(Abarbanell & Bonin, 1997).
In December 1995, the fourth commercial bank, Bank Gdañski (BG) was privatized
with a two-tier IPO: 33% of the shares to be sold to foreign portfolio investors using global
depository receipts and a 33% stake to domestic investors. A 4% stake was reserved to
employees and 30% to the state. Bank Inicjatyw Gospodarczch (BIG), a Polish private bank,
and two of its subsidiaries took a 24.1% stake in the domestic tranche and subsequently
increased this to 31% while the state retains ownership of 40% of the shares in BG
(Abarbanell & Bonin, 1997).
In October 1995, an “Outline of the Program of Consolidation and Privatization of
State-Owned Banks” was prepared, marking a shift in the privatization process. The emphasis
of the program was on two mergers of large banks. The main objective of the program was to
prevent a further foreign capital involvement in the privatization of financial institutions in
Poland. In fact, Polish banks were considered to be too small in size to survive the invasion
of foreign banks when entry restrictions are relaxed according to the association agreement
with the European Union (Bonin & Leven, 1996). However, due to a strong opposition of
managers of some state-owned banks, the program was revised. Instead of merging banks, in
July 1996 a decision was taken to form a banking group. Three commercial banks that had
been finally chosen- Bank Depozytowo Kredytowy (BDK), Powszechny Bank Gospodarczy
(PBG) and Pomorski Bank Kredytowy (PBKS)- formed a bank group together with PEKAO
-14-
S.A. The latter bank was to be a dominant partner for the other three subordinate banks. At
the same times, two banks - PBK and BH - prepared their own privatization plans. In the first
half of 1997, the minority of shares of PBK was sold to 3 domestic financial institutions with
a 39% stake altogether. The State Treasury retained over 50% of shares. In the case of BH,
the State Treasury sold 25.96% of shares to three foreign investors (J.P. Morgan, Swedbank
and Zurich Insurance Company) and 59% was sold by IPO. The State Treasury kept only
7.9% of votes at the general assembly of shareholders and 28−30% of shares via convertible
bonds (Balcerowicz & Bratkowski, 2001).
The second wave of privatization in Poland came after the election of the new pro-
reform coalition government in September 1997. The main concept of the privatization policy
adopted at that time was to choose reputable foreign strategic investors in order to achieve a
good governance structure in banks and receive capital and technology injections. An
additional aim was to collect substantial privatization revenues. On 1st January 1999, Poland
removed restrictions applied formerly to foreign banks concerning purchases of bigger stock
blocks, opening new branches and receiving a license to establish a bank in Poland. This
liberalization was a consequence of commitments undertaken by Poland when joining OECD.
Bank PEKAO was sold in 1998 by IPO (15%) and in 1999 to a strategic investor:
Uni-Credito Italiano (52.09% together with Allianz). The last-standing state-owned
commercial Bank Zachodni was sold to Allied Irish Banks with a 80% stake. The remaining
shares of the State Treasury in already partly privatized banks were sold to dominant
shareholders. Finally, PBK was taken over by Bank Austria Creditanstalt, while BH was
bought in 2000 by Citibank (Balcerowicz & Bratkowski, 2001).
Overall, a salient feature can be drawn from the privatization experiences of the three
countries as described above with regard to the role of foreign investors. Indeed, Hungary
was earliest and smooth in allowing entries of strategic foreign investors. As a result, they
ended up with the strongest banking system in the region. By contrast, although the Czech
Republic and Poland started the privatization processes very early (even earlier than
Hungary), political deterrence and a general defiance vis-à-vis foreign acquisitions of state-
owned banks due to fear of foreign control of domestic banking systems have considerably
retarded the transformation of banks into independent economic agents, and as a result the
improvement of banking efficiency.
-15-
3. Data
3.1. Sample For the greenfield entry sample, we wish to thank S. Claessens from the International
Monetary Fund for generously providing us with the data. His data is taken from Bureau van
Dijk's BankScope database, and originally run from 1987 to 1996. We first delete the years
1987 and 1996 due to their having too little observations. Because we control for reforms
other than liberalization to foreign entry, and data for these reforms are available only from
1989, the year 1988 is also naturally excluded from the sample. The final sample is
unbalanced, consisting of 9 banks at a minimum in 1989, and 75 banks at a maximum in
1994. Only commercial banks are taken into account. We would have liked to distinguish
between foreign and domestic banks, but the data doesn’t allow us to do this.
For the acquisition entry sample, we also obtained data from BankScope, version
2006. As it is commonly known by academic researchers and professionals, the data for
banks from less developed and transition countries require substantial editing before a
reliable sample can be constructed. We therefore carefully review our data to avoid double
counting of institutions, to exclude non-bank financial institutions from the sample. We also
exclude banks that are not commercial, cooperative and savings banks to ensure that we are
estimating performance based on a relatively homogeneous group of banks. With respect to
the type of account, we prefer unconsolidated accounts to consolidated ones wherever
available. Once the type of account chosen, we take the account following the international
financial reporting standards (IFRS) wherever available. If not, we take the account with
local accounting standards. Nevertheless, sometimes, IAS data are available for only one or
two years while longer time series are available in local standards. In such cases we use local
standards. Thus, our final sample is unbalanced, consisting of 34 banks at a minimum in
1994, and 98 banks at a maximum in 1999 and 2000. The period under study runs from 1994
to 2004. The distribution of banks by ownership and mode of entry is given in table 4.2.
While BankScope has a very large coverage of banks over the world, this database provides
information on bank ownership only for the current calendar year. Thus, we have had to track
the evolution in the ownership of each bank over time through several sources, including
banks’ official publications and Zephyr (Bureau Van Dijk)6. This enabled us to differentiate
6 We are grateful to R.d. Haas and I.v. Lelyveld from de Nederlandsche Bank for kindly sharing data on bank ownership for the years 1994 to 2001.
-16-
between foreign and domestic banks, and, amongst foreign banks, between greenfield
investments and merger and acquisition entries.
For both samples, we select the three most advanced countries in Central Europe, i.e.
the Czech Republic, Hungary, and Poland for reasons discussed in the introduction.
3.2. Variables Bank performance. To measure bank performance, we use two conventional indicators of the
well-being of firms: accounting measures of profitability and costs. Banks’ profits are
measured by the ratio of profits before tax over total assets. Banks’ costs are measured as
overhead over total assets. We also look at a specific measure of banking performance:
banks’ interest margins.
Liberalization Dates. We determine the dates on which a text was passed by legislation
allowing entry by greenfield banks, and purchases of domestic banks by foreign banks in the
Czech Republic, Hungary, and Poland. Choosing these dates is arduous, and sometimes
requires judgements. For the acquisition entry date, in case there are multiple such dates, we
choose that on which foreign banks were allowed to acquire a majority ownership stake in
domestic banks.
Indeed, in the Czech Republic, the Act on Banks was passed in February 1992, which
allowed the licensing of foreign banks. Therefore, we choose 1992 as the greenfield entry
date for the Czech Republic.
In Hungary, although foreign banks were already present before 1987, and the break-
up of the mono-bank system occurred in 1987, we do not choose this year as the greenfield
entry date because at this time, the country was still under centralized economy. In 1991, the
banking Act was passed, and the influx of foreign banks began in this year (see section 2.1).
Therefore, we choose 1991 as the greenfield entry date for Hungary.
In Poland, the Act on Banking voted in January 1989 effectively opened up the
market to foreign investors. Therefore, 1989 is the greenfield entry date for Poland.
We now discuss the choices of liberalization dates for the acquisition entry sample. In
the Czech Republic, there were two waves of privatization. The first wave of privatization by
voucher led to diffuse ownership in banks, with the government retaining a controlling stake,
and foreign participation being limited to a maximum of 25% stake. The second wave began
in 1997 with the adoption of a new privatization program aiming at accelerating state-
-17-
controlled banks’ privatization. Subsequently, foreign banks entered by acquiring a majority
ownership stake in partially privatized domestic banks. Although Zivnostenka Banka was
completely sold to foreign investors in 1992, we consider this as an exceptional case rather
than the beginning of a movement which can have impacts on the banking sector (see section
2.3). Therefore, we choose 1997 as the acquisition entry date for the Czech Republic.
For Hungary, things are much clearer as the privatization program adopted in 1994
was the unique program precluding a privatization process that quickly completed by the end
of 1997 (see section 2.3). Therefore, 1994 is the acquisition entry date for Hungary.
In Poland, as in the Czech Republic, there were two waves of privatization. The first
wave was initiated in 1991 when a program of privatization of state-owned banks was
adopted. It led to partial privatizations as foreign banks were allowed to participate with only
a small stake while the government still kept a substantial part in banks. The second wave
began in 1999 with the removal of restrictions applied formerly to foreign banks concerning
purchases of bigger stock blocks, thus enabling them to acquire a majority ownership stake in
domestic banks (see section 2.3). Therefore, 1999 is chosen as the acquisition entry date for
Poland.
Bank Characteristics. We take into account the following characteristics of banks: bank size,
as denoted by log of total assets; loans over total assets, which reflects a bank’s asset
structure7; equity ratio equal to equity over total assets, which reflects the risk preferences of
banks; operating costs as measured by overheads over total assets.
Banking Sector Structure Variables. We include three proxies for banking sector structure:
market share as measured by the share of a bank’s assets over the banking system’s total
assets; Herfindahl index as denoted by the sum of squares of banks’ market shares; and two
dummies for foreign takeover and domestic banks.
Other Reform Variables. To control for the possibility that the evolution in bank performance
measures is the results of other reforms rather than the liberalization to foreign bank entry, we
take into account indexes of banking, price and enterprise reforms obtained from EBRD.
7 One argument that could justify the control for the structure of assets can be drawn from (Havrylchyk, 2006) for the case of Poland “During the analyzed period the structure of assets underwent a drastic change. Previous easy sources of income, such as Treasury bonds and loans to blue-chip companies, were exhausted. The banks had to tap new and riskier segments of the market, such as households and small and medium enterprises».
-18-
Macroeconomic Environment Variables. We control for the following standard
macroeconomic environment characteristics as in a number of studies on operating
performance8: growth rate of real GDP per capita, inflation, and short-term interest rate of
deposits. Nevertheless, we don’t include these variables in the regressions on Greenfield
entry as the data available for the Czech Republic start only from 1992, which greatly reduce
the length of the period under study. Instead, we include country-fixed effects.
Table 1 summarizes variable definitions, and table 4 reports descriptive statistics of
the main variables. When one looks at the figures carefully, a few quick conclusions can be
drawn. First, contrary to common beliefs on transition economies, the average net interest
margins of banks in our samples are comparable to western standards (4,4% for the
Greenfield entry sample, and 3,7% for the acquisition entry sample). However, for 10% of
the banks in the Greenfield entry sample, the margins are over 8%, and for 1% of them, they
are over 10%. With regard to the acquisition entry sample, for 5% of the banks, the margins
are over 8% and for 1% of them, they are over 16%. Second, loans occupy a major part of the
overall portfolio, in the order of 42.17% for the Greenfield entry sample, and 44.9% for the
acquisition entry sample. Of course, we are aware of data problems that are relevant for all
transition countries. Therefore, when analyzing the findings of our research, we should keep
this bias in mind.
4. Results In this section, we discuss the results of the empirical tests. In the case of liberalization to
entry of greenfield banks, we examine the evolution in performance of the whole sample of
banks due to lack of data on ownership as discussed earlier. With regard to the entry of
takeover banks, we consider the evolution in performance of two sub-samples: the whole
sample of banks composed of domestic, greenfield, and takeover banks, and the sample of
domestic banks alone.
We find a significant drop in banks’ profits in the order of approximately 10%
following the liberalization to the entry of Greenfield banks (table 5). No significant effect is
observed on interest margins nor on overhead cost. The drop in bank profitability seems
large. However, we also observe a similar decrease of bank profitability in the order of 11%
due to other banking reforms. This might be explained by the fact that domestic banks may
have to modify their behavior in response to the entry of Greenfield banks by upgrading their
8 See, for example, (Claessens, Demirgüç-Kunt, & Huizinga, 2001), (Barajas, Steiner, & Salazar, 2000).
-19-
practices and technologies, which incurs non negligible cost. At the same time, attempts to
rationalize cost through reorganization and hardening of budget constraints take time to
implement. We don’t believe that the drop in bank profitability is due to increased
competition from greenfield banks because if this was the case, we would also observe a
decrease in interest margins. Moreover, stylized facts on transition banking often indicate
that, at the start of the transition, Greenfield banks operated on different market segments and
did not compete with domestic banks9.
With regard to the entry of foreign banks by acquiring domestic banks, bank
profitability has also dropped significantly following liberalization. The magnitude of the
decrease in profitability is greater for domestic banks than for the whole sample of banks
(2,5% vs. 1,6%). The reason might be that of the three types of banks, greenfield banks –
included in the whole sample – is least affected by liberalization. Indeed, we’ve run
regressions on the sample of solely greenfield banks, and found that the profitability of these
banks dropped by approximately 0.5%, but at a non-significant level10. These results are
consistent with a number of studies, such as (Claessens, Demirgüç-Kunt, & Huizinga, 2001)
for a pool of 80 countries, and (Denizer, 2000) for Turkey. In addition, (Clarke G. R., Cull,
D'Amato, & Molinari, 1999), in a study on Argentina, find decreased profitability only for
banks which engage in lending to manufacturing sectors, a market segment where foreign
banks are strongly present. (Unite & Sullivan, 2003) found the same result for Philippine
banks affiliated to a family group.
Looking at the results for interest margins in table 9, it is again evident that these have
dropped significantly following liberalization. Again, the magnitude of the change in interest
margins is greater for domestic banks than for the whole sample of banks (1.04% vs. 0.54%).
These results are in contrast with (Claessens, Demirgüç-Kunt, & Huizinga, 2001) and
(Denizer, 2000) which don’t find any significant impact of foreign bank entry on
intermediation costs. However, they are consistent with (Barajas, Steiner, & Salazar, 2000)
for Colombia, (Clarke G. R., Cull, D'Amato, & Molinari, 1999) for Argentina, and (Unite &
Sullivan, 2003) for the Philippines.
How to explain the significant drop in banks’ profitability and interest margins
following the removal of barriers to entry of foreign banks by acquiring domestic banks?
9 According to an account by (Abarbanell & Bonin, 1997, p. 36) for the case of Poland, the presence of foreign banks in the early years of transition has not engendered much competition for domestic banks. They have mainly followed their own clients into Poland, and generally limited their activities to servicing their clients’ banking needs. 10 Results are available from the author upon request.
-20-
Indeed, bank regulators may influence bank efficiency by affecting the degrees of current and
potential competition in banking markets with their policies on new bank formation and bank
mergers. In our case, the removal of legal barriers to entry of foreign banks first influences
the potential competition in the market because it creates a threat of entry by potential
competitors. It can also affect the degree of current competition if entries effectively take
place following liberalization, which is the case in the three countries in question in this
paper. Indeed, a look at the evolution of the Herfindahl index in each country does indicate an
intensified competition as these indices for the Czech Republic and Hungary bounced
exceptionally in 1996 and 1997 respectively, then dropped substantially, and stabilized from
1998 onwards. The index for Poland dropped constantly between 1994 and 1997, and then
stabilized. Competition dissipates monopoly profits of large commercial banks, and also
drives down interest margins.
Czech Republic
Hungary
Poland1000
1500
2000
2500
3000
Her
finda
hl In
dex
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
Year
Figure 1: Evolution of the Herfindahl Index (Source: author’s own calculation)
The results for operating costs are shown in table 4.10. It can be observed that cost
rises significantly for both samples following liberalization. Once again, the magnitude of the
increase in cost is more important for domestic banks than for the whole sample (3.54% vs.
2.17%). These results are in stark contrast with (Claessens, Demirgüç-Kunt, & Huizinga,
2001) and (Denizer, 2000), but consistent with (Clarke G. R., Cull, D'Amato, & Molinari,
-21-
1999) who finds a significant increase in administrative cost for banks concentrating on
mortgages lending where the presence of foreign banks is increasing.
How to explain the increase in operating costs following entry of takeover banks? In
principle, microeconomics and corporate governance theories would suggest drops in banks’
cost. Indeed, the threat of entry by potential stronger competitors from abroad following the
removal of legal barriers to foreign entry would force domestic bank managers to give up
their “quiet life” and to exert greater effort to reach cost efficiency. On the other hand, the
removal of legal barriers to the entry of foreign banks by acquiring domestic banks creates a
market for corporate control, which, according to corporate governance theories, may serve
as a means to discipline bank managers. In theory, one of the major benefits of corporate
takeovers is that the threat of a takeover provides management with the incentive to
maximize firm value. Otherwise, the firm may be acquired by another economic agent to be
transformed into a more efficient entity, and as a result, the current management is replaced.
The consequence of such a threat is that incentives exist for managers to efficiently run their
firms.
However, in our case, the decrease in cost resulting from disciplining effects on bank
managers may not be large enough to offset the increase in operating cost stemming from the
fact that merged and acquired banks need substantial restructuring which incurs non
negligible cost. Moreover, domestic banks may also need considerable new investment,
especially on information technology infrastructure, to withstand the challenge from foreign
competitors.
Turning into control variables, we observe that size is significantly and positively
related to profitability, but negatively related to operating cost. This might indicate some
scale effects which allows banks to achieve higher performance, and at the same time to
reduce cost. Banks with higher capital asset ratios tend to enjoy significantly higher
profitability and interest margins, which might be explained by the existence of depositor
discipline in transition banking. Indeed, well-capitalized firms face lower expected
bankruptcy cost for themselves and for their customers, thus can get a reduction of their cost
of funding (Berger A. N., 1995). This represents a form of depositor market discipline which
has been confirmed by empirical evidence (Goldberg & Hudgins, 2002) and (Park &
Peristiani, 1998) for the U.S. savings and loan associations, and by (Karas, Pyle, & Schoors,
2006) for Russia. (Martinez Peria & Schmukler, 1999) finds evidence that market discipline
more generally exists in developing countries, even in the presence of deposit insurance. In
-22-
Central and Eastern Europe, the recurrent asset quality problems and bank failures episodes
would induce depositors to act prudently and to avoid deposits in under-capitalized banks.
Portfolio composition, as reflected by the loan variable, is significantly and positively
related to net interest margins in both the Greenfield and acquisition entry cases, which
indicates that banks with a higher proportion of loans in their portfolio tend to have higher
margins. Since loans are often the most risky asset11, this finding indicates that banks have
integrated the risk factor in their loan pricing, and consequently, lending results in wider
margins. Operating costs tend to reduce significantly domestic banks’ profitability in the
acquisition entry sample, but increase significantly interest margins in both entry cases. This
lends support to the efficient-structure hypothesis which suggests that differences in banks’
profitability and interest margins are attributable, in opposite ways, to differences in
operational efficiency.
Concerning banking structure variables, M&As and domestic banks tend to have
lower profitability than Greenfield banks, which is consistent with existing findings in the
literature pointing to a superior performance of Greenfield banks. The negative coefficients
on both merger and domestic in the net interest margin regression indicate that these types of
banks also experience lower interest margins than Greenfield banks. These results
corroborate previous results by (Demirguc-Kunt & Huizinga, 1999) (Barajas, Steiner, &
Salazar, 2000), and might reflect the fact that the better reputation and superior banking
expertise and services of Greenfield banks may allow them to borrow funds at lower rates
while at the same time lending at higher rates than their counterparts.
The coefficients on herfin are positively and significantly related to profitability and
interest margins in the Greenfield entry case, a finding that lends support to the traditional
structure-conduct-performance (SCP) hypothesis according to which banks in more
concentrated markets earn higher profitability and interest margins due to a non-competitive
pricing behavior. However, the Herfindahl index is negatively and significantly related to
both profitability and interest margin in the acquisition entry sample, although the magnitude
of the impact seems very small. This result, though unexpected, corroborates those obtained
by (Claeys & Vander Vennet, 2004) for a sample of Eastern European countries, and might
be a reflection of the rapid development of bank lending in these countries in the latter part of
the 1990s. Market share is negatively related to profitability and interest margin for the whole 11 (Havrylchyk, 2006) offers indications of augmented risks in lending for the case of Poland “During the analyzed period the structure of assets underwent a drastic change. Previous easy sources of income, such as Treasury bonds and loans to blue-chip companies, were exhausted. The banks had to tap new and riskier segments of the market, such as households and small and medium enterprises».
-23-
sample of banks in the acquisition entry sample. This result does not corroborate the relative-
market-power hypothesis12 and means that ceteris paribus, larger banks are not able to reap
the benefices of their market power in pricing.
Amongst other reforms, enterprise reform appears to exert most significant impact by
increasing banks’ profitability, and reducing margins in the acquisition entry sample. With
respect to the macroeconomic environment, inflation and interest rates affect profitability and
interest margins positively. We don’t observe any effect of the business cycle as denoted by
the growth variable on bank performance.
In summary, the liberalization to entry of greenfield banks has had a limited effect on
bank performance. By contrast, bank performance has significantly improved following the
liberalization to foreign mergers and acquisitions of domestic banks. These results therefore
confirm the differential implications of the two types of foreign entry as suggested by
theoretical analysis on foreign bank entry.
5. Conclusions This paper has investigated the effects of the two modes of entry of foreign banks –
Greenfield investments vs. mergers and acquisitions of domestic banks - in an empirical
setting. The results show a very limited effect of the entry of Greenfield banks on domestic
banking market in the early transition period. In contrast, the foreign entry by mergers and
acquisitions of domestic banks exerts significant impacts on bank performance. Indeed, we
observe significant declines in banks' profits and net interest margins, and a significant
increase in operating costs. These results therefore confirm the differential implications of the
two types of foreign entry as suggested by theoretical analysis on foreign entry.
Our results have important policy implications. Indeed, they show that the
liberalization to the entry of de novo banks is not sufficient to bring significant changes to the
domestic banking market. Real benefits come mainly with foreign acquisitions of domestic
banks. These conclusions contribute to shed light on the decisions to be made by those
emerging and transition countries still hesitant to engage in this way.
The policy implications of our analysis should, nevertheless, be treated with caution.
We have investigated the effect of the liberalization to entry of de novo banks in a period
where the scope of activities of these banks was rather limited. This may bias our results. It
12 This theory suggests that banks with higher market shares are able to exercise power in pricing, and consequently earn higher market share.
-24-
would be preferable to extend the period studied into the one in which Greenfield banks
started being interested in looking for local business opportunities.
-25-
References [to be completed]
Abarbanell, J. S., & Bonin, J. P. (1997). Bank Privatization in Poland: The Case of Bank
Slaski. Journal of Comparative Economics , 25 (1).
Balcerowicz, E., & Bratkowski, A. (2001). Restructuring and Development of the Banking
Sector in Poland. Lessons to be Learnt by Less Advanced Transition Countries. Center for
Social and Economic Research Report , 44.
Barajas, A., Steiner, R., & Salazar, N. (2000). The Impact of Liberalization and Foreign
Investment in Colombia's Financial Sector. Journal of Development Economics , 63, 157-
196.
Bekaert, G., & Harvey, C. R. (2004). Chronology of Important Financial, Economic and
Political Events in Emerging Markets.
http://www.duke.edu/~charvey/Country_risk/chronology/bibliography.htm .
Berger, A. N. (1995). The Relationship Between Capital and Earnings in Banking. Journal of
Money, Credit and Banking , 27, 432-456.
Berger, A. N., DeYoung, R., Genay, H., & Udell, G. F. (2000). Globalization of financial
institutions: evidence from cross-border banking performance. Brookings-Wharton Papers
on Financial Services .
Bonin, J. P., & Leven, B. (1996). Polish Bank Consolidation and Foreign Competition:
Creating a Market-Oriented Banking Sector. Journal of Comparative Economics , 23 (1).
Bonin, J. P., Hasan, I., & Wachtel, P. (2005b). Bank Performance, Efficiency and Ownership
in Transition Countries. Journal of Banking and Finance , 29, 31-53.
Bonin, J., & Wachtel, P. (1999). Towards Market-Oriented Banking in the Economies in
Transition. In M. Blejer, & M. Skreb, Financial Sector Transformation: Lessons from the
Economies in Transition. Cambridge: Cambridge University Press.
Buch, C. (2003). Information or regulation: What drives the international activities of
commercial banks? Journal of Money, Credit, and Banking , 35 (6), 851–869.
Buch, C. M. (2003). Information or regulation: What drives the international activities of
commercial banks? Journal of Money, Credit, and Banking , 35 (6), 851–869.
Claessens, S., Demirgüç-Kunt, A., & Huizinga, H. (2001). How does foreign entry affect
domestic banking markets? Journal of Banking and Finance , 25 (5), 891-911.
Claeys, S., & Hainz, C. (2006). Acquisition versus Greenfield: The Impact of the Mode of
Foreign Bank Entry on Information and Bank Lending Rates. European Central Bank
Working Paper Series .
-26-
Claeys, S., & Vander Vennet, R. (2004). Determinants of Bank Interest Margins in Central
and Eastern Europe: A Comparison with the West. http://www.eu-financial-
system.org/fileadmin/content/Dokumente_Events/first_conference/Claeys_VanderVennet.p
df .
Clarke, G. R., Cull, R., D'Amato, L., & Molinari, A. (1999, August). The Effect of Foreign
Entry on Argentina's Domestic Banking Sector. The World Bank Policy Research Working
Paper Series .
Demirguc-Kunt, A., & Huizinga, H. (1999). Determinants of Commercial Bank Interest
Margins and Profitability: Some International Evidence. World Bank Economic Review ,
13 (2), 379-408.
Denizer, C. (2000, October). Foreign entry in Turkey's banking sector, 1980-97. World Bank
Policy Research Working Paper Series .
Goldberg, L., & Hudgins, S. (2002). Depositor Discipline and Changing Strategies for
Regulating Thrift Institutions. Journal of Financial Economics , 63, 263–274.
Hauswald, R., & Marquez, R. (2003). Information technology and financial services
competition. Review of Financial Studies , 16 (3), 921–948.
Havrylchyk, O. (2006). Efficiency of the Polish banking industry: Foreign versus domestic
banks. Journal of Banking and Finance , 30 (7), 1975-1996.
Kaas, L. (2004). Financial market integration and loan competition—When is entry
deregulation socially beneficial? European Central bank Working Paper Series .
Karas, A., Pyle, W., & Schoors, K. (2006). Sophisticated Discipline in a Nascent Deposit
Market: Evidence from Post-Communist Russia. BOFIT Discussion Papers , 13.
Kormendi, R. C., & Schnatterly, K. (1996). Bank Privatization in Hungary and the Magyar
Kulkereskedelmi Bank Transaction. William Davidson Institute Working Papers Series .
Lehner, M., & Schnitzer, M. (2008). Entry of foreign banks and their impact on host
countries. Journal of Comparative Economics , doi:10.1016/j.jce.2008.02.002.
Majnoni, G., Shankar, R., & Varhegyi, E. (2003). The Dynamics of Foreign Bank Ownership
- Evidence from Hungary. The World Bank Policy Research Working Paper Series , 3114.
Martinez Peria, S., & Schmukler, S. (1999). Do Depositors Punish Banks for Bad Behavior?
Market Discipline in Argentina, Chile, and Mexico. World Bank Policy Research Working
Paper , 2058.
Mathieson, D. J., & Schinasi, G. J. (2000). World Economic and Financial Surveys:
International Capital Markets: Developments, Prospects, and Key Policy Issues.
International Monetary Fund.
-27-
Park, S., & Peristiani, S. (1998). Market Discipline by Thrift Depositors. Journal of Money,
Credit, and Banking , 30, 347–364.
Pruteanu-Podpiera, A., Weill, L., & Schobert, F. (2008). Banking Competition and
Efficiency: A Micro-Data Analysis on the Czech Banking Industry. Comparative
Economic Studies , forthcoming.
Sengupta, R. (2007). Foreign entry and bank competition. Journal of Financial Economics ,
84 (2), 502-528.
Tang, H., Zoli, E., & Klytchnikova, I. (2000, November). Banking Crises in Transition
Economies: Fiscal Costs and Related Issues. World Bank Policy Research Working Paper,
2484.
Tuma, Z. (2002). Banking Sector Development in the Czech Republic. Mimeo .
Van Tassel, E., & Vishwasrao, S. (2007). Asymmetric information and the mode of entry in
foreign credit markets. Journal of Banking and Finance , 31 (2), 3742-3760.
Vo Thi, N. A., & Vencappa, D. (2008, August). Does the Entry Mode of Foreign Bank
Matter for Bank Efficiency? Evidence from Transition Countries. The William Davidson
Institute Working Paper Series , 925.
Weill, L. (2003). Banking Efficiency in Transition Economies: The Role of Foreign
Ownership. Economics of Transition , 11, 569-592.
-28-
Tableau 1: Definitions of variables
Name Definition Source of data Dependent variables profit ratio of profits before tax to total assets; in percentage. It
denotes a bank’s profitability. BankScope
irmargin ratio of net interest revenue to total assets; in percentage. It denotes a bank’s net interest margin.
BankScope
oper_cost ratio of overheads to total assets; in percentage. It denotes a bank’s operating costs.
BankScope
Liberalization variables greenfield dummy variable indicating liberalization to entry of
Greenfield banks; 1 for the years following the liberalization, and 0 otherwise.
Author's own research
takeover dummy variable indicating liberalization to foreign acquisition of domestic banks; 1 for the years following liberalization, and 0 otherwise.
Author's own research
Bank-specific variables size log of total assets. It denotes bank size. BankScope equity ratio of equity to total assets; in percentage. It denotes the
risk preferences of banks. BankScope
loan ratio of loans to total assets; in percentage. It denotes a bank’s portfolio composition.
BankScope
Banking sector structure variables merger dummy variable for foreign takeover bank. R.d. Haas, I.v. Lelyveld &
author's own research domestic dummy variable for domestic bank. R.d. Haas, I.v. Lelyveld &
author's own research herfin Herfindahl index; sum of squares of banks’ market shares.
It is an index of market concentration. BankScope
sharemar ratio of a bank’ assets to total banking system’s assets; in percentage. It denotes a bank’s market share of assets.
BankScope
Reform variables bank index of banking reform. EBRD enter index of enterprise reform. EBRD price index of price liberalization. EBRD Country-specific control variables lgdp log of real gdp per capita. PWT 6.2 inflation annual inflation of the GDP deflator. WDI irate for the Czech Republic and Hungary: average rate
weighted by volume on the three-month Treasury bills sold at auctions (Line 60C.ZF); for Poland: discount rate (end of period) (Line 60.ZF).
IFS
-29-
Tableau 2: Distribution of banks by ownership and modes of entry, Acquisition entry sample
Czech Republic Hungary Poland Total 1994
domestic 6 5 13 24 greenfield 5 3 2 10
mergers Total 11 8 15 34
1995 domestic 6 6 14 26
greenfield 5 4 3 12 mergers
Total 11 10 17 38 1996
domestic 9 6 20 35 greenfield 5 6 3 14
mergers 1 1 Total 14 13 23 50
1997 domestic 9 6 28 43
greenfield 10 6 6 22 mergers 3 3 6
Total 19 15 37 71 1998
domestic 11 7 26 44 greenfield 11 10 13 34
mergers 2 9 3 14 Total 24 26 42 92
1999 domestic 11 6 26 43
greenfield 12 13 14 39 mergers 4 9 3 16
Total 27 28 43 98 2000
domestic 10 6 21 37 greenfield 12 14 14 40
mergers 5 9 7 21 Total 27 29 42 98
2001 domestic 8 7 15 30
greenfield 11 11 16 38 mergers 7 8 8 23
Total 26 26 39 91 2002
domestic 6 9 12 27 greenfield 10 12 17 39
mergers 9 8 10 27 Total 25 29 39 93
2003 domestic 6 8 13 27
greenfield 10 12 17 39 mergers 9 7 12 28
Total 25 27 42 94 2004
domestic 6 7 12 25 greenfield 10 14 16 40
mergers 8 7 12 27 Total 24 28 40 92
-30-
Tableau 3: Liberalization dates
Country Liberalization Type Year Greenfield Entry 1992 Czech Republic Acquisition Entry 1997 Greenfield Entry 1991 Hungary Acquisition Entry 1994 Greenfield Entry 1989 Poland Acquisition Entry 1999
-31-
Tableau 4: Descriptive statistics
Greenfield Entry Sample*
Variable Observations Mean S.D Min Maxprofit 311 1.91 5.21 -49.69 12.21irmargin 310 4.40 2.77 -8.23 14.76lnassets 315 8.87 2.20 2.42 12.87loan 315 42.17 19.64 0.00 94.75equity 315 13.08 12.82 -3.47 82.48sharemar 315 6.67 11.90 0.03 100.00oper_cost 292 4.14 7.42 0.07 99.30
Acquisition Entry Sample*
Variables Observations Mean S. D Min Max Wh. Sample Domestic Wh. Sample Domestic Wh. Sample Domestic Wh. Sample Domestic Wh. Sample Domestic profit 859 360 0.80 0.06 4.35 6.05 -39.18 -39.18 14.52 14.52 irmargin 857 360 3.72 3.80 2.76 2.48 -8.06 -8.06 21.56 21.56 lnassets 861 361 6.52 6.58 1.48 1.53 1.71 2.48 10.29 10.29 loan 861 361 44.90 42.11 20.62 19.73 0.00 0.00 98.01 96.91 equity 861 361 12.68 11.80 13.53 14.54 -55.56 -55.56 98.41 84.47 sharemar 861 361 3.83 5.08 6.34 8.18 0.01 0.04 51.93 51.93 oper_cost 856 359 4.63 5.07 5.79 5.06 -1.88 -1.88 129.17 60.63
*All variables are in percentage, except lnassets.
-32-
Tableau 5: Changes in Bank Profitability following Greenfield Entry
Independent variables Whole sample greenfield -9.605 (3.643)*** size 0.811 (0.518) loan 0.024 (0.015) equity 0.159 (0.055)*** sharemar 0.015 (0.039) oper_cost 0.041 (0.060) herfin 0.004 (0.002)** bank -11.814 (4.644)** enter -2.075 (1.173)* price 0.526 (0.507) country_2 2.221 (2.398) country_3 10.205 (3.355)*** year_2 -46.416 (19.399)** year_3 -40.393 (16.915)** year_4 -26.266 (10.745)** year_5 -14.641 (5.823)** year_6 -1.620 (0.890)* year_7 0.424 (0.611) Constant 30.403 (14.557)** Observations 287 Number of banks 77 R-squared 0.19 NOTES: Panel data econometrics techniques are used in doing regressions. Indeed, an Hausman test indicates that a random-effects specification is appropriate. A Breusch-Pagan test for group-wise heteroskedasticity, and the Wooldridge (2002) test for serial correlation indicate the presence of heteroskedasticityin the residuals. We therefore choose the random-effects regression with robust standard errors. Sample period: 1989-1995. Robust standard errors are in parentheses. * denotes a significant level at 10% level; ** at 5%; and *** at 1%.
-33-
Tableau 6: Changes in Banks' Net Interest Margin following Greenfield Entry
Independent variables Whole sample greenfield -1.359 (1.902) size 0.467 (0.160)*** loan 0.034 (0.007)*** equity 0.029 (0.019) sharemar -0.044 (0.021)** oper_cost 0.140 (0.027)*** herfin 0.002 (0.001)** bank -3.054 (2.144) enter -1.045 (0.670) price 0.516 (0.274)* country_2 2.943 (0.805)*** country_3 5.999 (0.962)*** year_2 -14.459 (9.073) year_3 -12.690 (8.001) year_4 -7.484 (5.169) year_5 -4.004 (2.844) year_6 -0.638 (0.468) year_7 0.003 (0.275) Constant 3.931 (8.417) Observations 286 Number of banks 77 R-squared 0.40 NOTES: Panel data econometrics techniques are used in doing regressions. Indeed, an Hausman test indicates that a random-effects specification is appropriate. A Breusch-Pagan test for group-wise heteroskedasticity, and the Wooldridge (2002) test for serial correlation indicate the presence of both
-34-
heteroskedasticity and autocorrelation of order 1 in the residuals. We therefore choose the Prais-Winsten regression with heteroskedastic panels corrected standard errors and an AR(1) process in the error terms. Sample period: 1989-1995. Robust standard errors are in parentheses. * denotes a significant level at 10% level; ** at 5%; and *** at 1%. Tableau 7: Changes in Banks' Operating Costs following Greenfield Entry
Independent variables Whole sample greenfield -3.021 (3.833) size -0.691 (0.470) loan -0.022 (0.017) equity 0.226 (0.127)* sharemar 0.107 (0.059)* herfin 0.002 (0.002) bank -5.618 (4.381) enter -2.938 (2.425) price -0.730 (0.523) country_2 5.782 (2.259)** country_3 0.239 (2.528) year_2 -34.975 (20.762)* year_3 -29.742 (18.081)* year_4 -18.866 (11.736) year_5 -11.539 (6.872)* year_6 -2.356 (1.367)* year_7 -0.949 (0.734) Constant 36.193 (18.489)* Observations 287 Number of banks 77 R-squared 0.20 NOTES: Panel data econometrics techniques are used in doing regressions. Indeed, an Hausman test indicates that a random-effects specification is appropriate. A Breusch-Pagan test for group-wise heteroskedasticity, and the Wooldridge (2002) test for serial correlation indicate the presence of both heteroskedasticity and autocorrelation of order 1 in the residuals. We therefore choose the Prais-Winsten
-35-
regression with heteroskedastic panels corrected standard errors and an AR(1) process in the error terms. Sample period: 1989-1995. Robust standard errors are in parentheses. * denotes a significant level at 10% level; ** at 5%; and *** at 1%.
-36-
Tableau 8: Changes in Banks' Profitability Following Liberalization to Foreign Acquisition of Domestic Banks
(1) (2) Independent variables Whole sample Domestic banks takeover -1.619 -2.535 (0.640)*** (1.266)** size 3.277 1.944 (0.956)*** (1.329) loan 0.006 -0.021 (0.015) (0.024) equity 0.176 0.150 (0.043)*** (0.056)*** sharemar -0.094 0.029 (0.055)* (0.068) oper_cost -0.204 -0.635 (0.158) (0.085)*** merger -2.873 (1.009)*** domestic -3.450 (0.596)*** herfin -0.003 -0.005 (0.001)*** (0.001)*** bank -0.361 0.178 (1.108) (2.188) enter 3.641 11.547 (1.856)** (3.456)*** price 10.453 14.971 (4.365)** (7.486)** growth 0.211 0.316 (0.183) (0.354) inflation 0.306 0.525 (0.073)*** (0.148)*** irate 0.089 0.294 (0.061) (0.100)*** Observations 854 368 Number of banks 148 77 R-squared 0.34 0.60 NOTES: Panel data econometrics techniques are used in doing regressions. Indeed, a Hausman test indicates that a fixed-effects specification is appropriate for both cases. A modified Wald test for group-wise heteroskedasticity, and the Wooldridge (2002) test for serial correlation indicate the presence of both heteroskedasticity and autocorrelation of order 1 in the residuals in both cases. We therefore choose the fixed-effects models with Newey-West standard errors and an AR (1) process in the error terms for both samples. Year dummies are included in the regressions but not reported here. Country dummies are also included but dropped due to colinearity. Sample period: 1994-2004. Robust standard errors are in parentheses. * denotes a significant level at 10% level; ** at 5%; and *** at 1%.
-37-
Tableau 9: Changes in Banks' Interest Margin Following Liberalization to Foreign Acquisition of Domestic Banks
(1) (2) Independent variables Whole sample Domestic banks takeover -0.539 -1.037 (0.333)* (0.518)** size 0.541 0.116 (0.288)* (0.132) loan 0.018 0.021 (0.006)*** (0.008)*** equity 0.092 0.081 (0.016)*** (0.016)*** sharemar -0.049 0.015 (0.020)** (0.020) oper_cost 0.100 0.067 (0.024)*** (0.022)*** merger -3.861 (1.332)*** domestic -3.767 (1.318)*** herfin -0.000 -0.001 (0.000)* (0.000)*** bank -0.258 -0.557 (0.497) (0.715) enter -2.451 -2.540 (0.826)*** (1.544)* price -1.367 0.154 (1.664) (2.049) growth 0.020 0.019 (0.078) (0.120) inflation 0.011 0.144 (0.036) (0.041)*** irate -0.041 0.053 (0.027) (0.032)* Constant 8.621 (8.025) Observations 852 369 Number of banks 147 78 R-squared 0.36 0.46 NOTES: Panel data econometrics techniques are used in doing regressions. Indeed, an Hausman test indicates that a fixed-effects specification is appropriate for (1), and a random-effects specification for (2). A modified Wald test for group-wise heteroskedasticity for (1), a Breusch-Pagan test for (2), and the Wooldridge (2002) test for serial correlation for both indicate the presence of both heteroskedasticity and autocorrelation of order 1 in the residuals in both cases. We therefore choose the fixed-effects model with Newey-West standard errors and an AR (1) process in the error terms for (1), and the Prais-Winsten regression with heteroskedastic panels corrected standard errors and an AR(1) process in the error terms for (2). Year dummies are included in the regressions but not reported here. Country dummies are also
-38-
included but dropped due to colinearity. Sample period: 1994-2004. Robust standard errors are in parentheses. * denotes a significant level at 10% level; ** at 5%; and *** at 1%.
-39-
Tableau 10: Changes in Banks' Operating Costs Following Liberalization to Foreign Acquisition of Domestic Banks
(1) (2) Independent variables Whole sample Domestic banks takeover 2.168 3.544 (1.316)* (1.845)** size -3.470 -1.222 (0.651)*** (0.325)*** loan -0.036 -0.044 (0.024) (0.023)* equity 0.013 -0.037 (0.085) (0.065) sharemar 0.091 0.037 (0.066) (0.051) merger -1.000 (1.290) domestic -0.935 (0.898) herfin 0.002 0.002 (0.001)* (0.001)** bank -2.973 -1.833 (2.711) (2.091) enter -3.870 5.097 (4.241) (5.666) price -17.090 -8.269 (11.174) (6.498) growth -0.408 0.191 (0.326) (0.330) inflation -0.077 -0.012 (0.100) (0.059) irate -0.061 0.116 (0.138) (0.109) Constant 34.351 (31.838) Observations 854 369 Number of banks 148 78 R-squared 0.13 0.14 NOTES: Panel data econometrics techniques are used in doing regressions. Indeed, an Hausman test indicates that a fixed-effects specification is appropriate for (1), and a random-effects specification for (2). A modified Wald test for group-wise heteroskedasticity for (1), a Breusch-Pagan test for (2), and the Wooldridge (2002) test for serial correlation for both indicate the presence of heteroskedasticity in both cases, and of autocorrelation of order 1 in the residuals in (1). We therefore choose the fixed-effects model with Newey-West standard errors and an AR (1) process in the error terms for (1), and the random-effects model with robust standard errors for (2). Year dummies are included in the regressions but not reported here. Country dummies are also included but dropped due to colinearity. Sample period: 1994-2004. Robust standard errors are in parentheses. * denotes a significant level at 10% level; ** at 5%; and *** at 1%.
DAVIDSON INSTITUTE WORKING PAPER SERIES - Most Recent Papers The entire Working Paper Series may be downloaded free of charge at: www.wdi.umich.edu
CURRENT AS OF 2/3/09
Publication Authors Date
No. 948: Banking Market Liberalization and Bank Performance: the Role
of Entry Modes Ngoc-Anh Vo Thi Jan 2009
No. 947: The Monetary Union: The Decade Ahead. The Case of Non-Member States
Daniel Daianu and Laurian Lungu Jan 2009
No. 946: Currency Substitution: A Case Of Kazakhstan (2000:1-2007:12) Mesut Yilmaz, Yessengali Oskenbayev & Kanat Abdulla
Jan 2009
No. 945: Determinants Of Pollution Abatement And Control Expenditure: Evidence From Romania
Guglielmo Caporale, Christophe Rault, Robert Sova & Anamaria Sova
Jan 2009
No. 944: Bootstrap panel Granger-causality between government spending and revenue in the EU
Antonio Afonso and Christophe Rault
Jan 2009
No. 943: Regional Determinants of FDI Distribution in Poland Agnieszka Chidlow and Stephen Young
Nov 2008
No. 942: Structural Reform and Firm Exports Alvaro Cuervo-Cazurra & Luis Alfonso Dau
Sept 2008
No. 941: Exploring The Relationship Between Military Spending & Human Rights Performance In South Asia
Krishna Chaitanya, Vadlamannati and K K Shakya Lahiru Pathmalal
Oct 2008
No. 940: Structural Reform And Firm Profitability In Developing Countries
Alvaro Cuervo-Cazurra & Luis Alfonso Dau
July 2008
No. 939: Does Timing Of Elections Instigate Riots? A Subnational Study Of 16 Indian States, 1958 – 2004
Krishna Chaitanya Vadlamannati Oct 2008
No. 938: Price Setting And Market Structure: An Empirical Analysis Of Micro Data
Fabrizio Coricelli & Roman Horvath Sept 2008
No. 937: Inflation Differentials in EU New Member States: An Empirical Evidence
Roman Horvath & Kamila Koprnicka Oct 2008
No. 936: Market Reforms and Growth in Post-socialist Economies: Evidence from Panel Cointegration and Equilibrium Correction Model
Igor Pelipas and Alexander Chubrik Sept 2008
No. 935: Derivatives trading and the volume-volatility link in the Indian Stock Market
S. Bhaumiky, M. Karanasosy and A. Kartsaklas
April 2008
No. 934: Yuan Real Exchange Rate Undervaluation, 1997-2006. How Much, How Often? Not Much, Not Often
Jeff Chen, Wende Deng & David Kemme
Aug 2008
No. 933: How Important Are Foreign Shocks in Small Open Economy? The Case of Slovakia
Roman Horváth and Marek Rusnák
Sept 2008
No. 932: Estimating Poverty for Indigenous Groups in Chile by Matching Census and Survey Data
Claudio A. Agostini, Philip H. Brown, and Andrei Roman
Aug 2008
No. 931: Is There Electoral Cycles In Globalization Process? Evidence From 78 Democratic Countries, 1975 – 2006
Krishna Chaitanya Vadlamannati Sept 2008
No. 930: Lobbying, Corruption & Other Banes Nauro Campos & Francesco Giovannoni
Sept 2008
No. 929: Do Elections Slow Down Economic Globalization Process In India? It’s Politics Stupid!
Krishna C Vadlamannati Aug 2008
No. 928: Impact Of Institutional Quality On Human Rights Abuses In Transition Economies
Krishna C Vadlamannati & Artur Tamazian
July 2008
No. 927: Do Choice & Speed Of Reforms Matter For Human Rights During Transition?
Krishna Chaitanya Vadlamannati July 2008
No. 926: Socioeconomic, Institutional & Political Determinants Of Human Rights Abuses: A Subnational Study Of India, 1993 – 2002
Krishna Chaitanya Vadlamannati July 2008
No. 925: Does the Entry Mode of Foreign Banks Matter for Bank Efficiency? Evidence from the Czech Republic, Hungary, and Poland
Ngoc-Anh Vo Thi & Dev Vencappa
July 2008